Diagnosis of aneurysm of the thoracic aorta. Comparison between two non invasive techniques: two-dimensional echocardiography and computed tomography

Transcription

1 European Heart Journal (1984) 5, Diagnosis of aneurysm of the thoracic aorta. Comparison between two non invasive techniques: two-dimensional echocardiography and computed tomography S. ILICETO*, G. ErroRREf, G. FRANCiosof, G. ANTONELLI*, G. BIASCO* AND P. RIZZON* *Division of Cardiology and the flnstitute ofradiology, University ofbari, Bari, Italy KEY WORDS: Aortic aneurysm, computed tomography, cross-sectional echocardiography. In order to assess the respective values of two-dimensional echocardiography (2D echo) and computed tomography (CT) in the evaluation of aneurysms of the thoracic aorta, 14 patients with angiographically proven aneurysms of the thoracic aorta (three of which were dissecting aneurysms) were studied. The entire thoracic aorta was visualized in 10/14 patients by 2D Echo and in all patients by CT. An intimal flap was recognized by 2D echo in each case with a dissection whereas such a recognition was never possible with CT. CT identified calcification of the wall of an huge aneurysm of the ascending aorta in one case and a thrombotic stratification in the lumen of the descending thoracic aorta in another case; both abnormalities were missed by echocardiography probably because of inappropriate gain setting. In conclusion, 2D Echo and CT are both useful in the evaluation of aneurysms of the thoracic aorta; 2D echo appears to be superior in the recognition of an intimal flap due to dissection whereas CT allows a better recognition of the configuration, extension and tissue modifications of the aneurysm. Cineangiography is generally accepted as the 'gold standard' for the diagnosis of aneurysms of the aorta because it enables an accurate visualization of the entire thoracic aorta, the identification of an intimal flap and the delineation of the extent of a dissection!']. Aortic aneurysms, with or without dissections, can also be diagnosed non-invasively: twodimensional echocardiography (2D-echo)I 2 " 4 ] and computed tomography (CT)! 5 ' 6 ] are the two noninvasive techniques most utilized for this diagnosis. The purpose of this study is to compare the diagnostic value of these techniques in a series of patients with angiographically proven aneurysms of the thoracic aorta. Methods Fourteen consecutive patients with angiographi- Received for publication on 12 May 1983 and in revised form 23 February Address for correspondence: Sabino Iliceto, M.D., Division of Cardiology, University of Bari-Policlinico, Bari, Italy cally proven aneurysms of the thoracic aorta were studied by 2D echo and CT. An aneurysm was defined angiographically as a localized dilatation of a portion of the thoracic aorta. The angiographic diagnosis of a dissection was based on the recognition of an intimal tear, identified as a radiolucent linear structure changing its position during the cardiac cycle. Attention was paid to the extent of the dissection in order to classify the patients with dissection according to the De Bakey classification: a type I dissection begins in the ascending aorta and extends beyond the arch; a type II dissection is confined to the ascending aorta; a type HI dissection originates in the descending thoracic aorta and propagates distal ly for a variable distance. A dissection was present in three patients (cases 1-3); in two (cases 1 and 2) it was a type I dissection, in the remaining patient (case 3) a type III dissection. In one patient (case 7) with a huge aneurysm of the ascending aorta the angiographic study showed heavy calcification of the aortic wall whose presence was confirmed during cardiac surgery. In another patient (case 11) there was a conspicu- 0I95-668X/84/ I J02.00/ The European Society of Cardiology

2 546 S. Iliceto el al. Table 1 Patient data Pts Age Aneurysm extension Etiology Aortic portion imaged by echo Notes 1 43 Hypertension Type I dissection* 2 71 Hypertension Type I dissection* DTA 26 Asc. aorta 30 Asc. aorta 65 Asc. aorta 51 Asc. aorta Hypertension Marfan syndr. Marfan syndr. Syphilis Syphilis 61 Hypertension 63 Hypertension 51 DTA Hypertension 65 DTA Hypertension Asc. aorta; arch Asc. aorta; DTA Asc. aorta; DTA Asc. aorta; DTA Asc. aorta; DTA Asc. aorta; arch Marfan syndr Asc. aorta Hypertension Hypertension Pts=patients. Asc.=ascending; DTA=descending thoracic aorta. Syndr. = syndrome. *Type of dissection according to De Bakey classification. ous thrombotic stratification in the enlarged lumen of the descending thoracic aorta (DTA). Twelve echocardiographic studies and 8 CT studies were performed and interpreted prior to angiography, the remaining studies were performed after the angiographic results were known. All CT studies were interpreted by one of us (G.E.); the echocardiographic studies were performed by two of us (S.I. and G.A.). In 7 cases the final echocardiographic diagnosis was made by agreement between S.I. and G.A. (only in one case there was disagreement), the remaining studies were performed and interpreted by S.I. Three patients had a Marfan syndrome, 9 had a history of severe hypertension, 2 had a luetic aneurysm. The patients' data are summarized in Table 1. Type 111 dissection* ; calcification of aortic walls (surgical confirmation) ; thrombotic stratification in DTA lumen (confirmed at autopsy) TWO-DIMENSIONAL ECHOCARDIOGRAPHY Two-dimensional echocardiographic studies were performed with a wide-angle phased array sector scanner (Toshiba SL 53M or SSH 40A, Hewlett Packard 77020A) with the patients in the supine position or in the left lateral decubitus, utilizing all the available tomographic planes obtainable from different approaches (precordial, apical, subcostal and suprastemal). Ascending aorta: the aortic root, the aortic valve and the ascending aorta were imaged in their long axis as well as in the short axis utilizing different tomographic planes obtainable from the precordial, apical and suprastemal approach (Figs 1A, 2, 3). The best definition of the aortic walls was achieved with the precordial long axis and short axis views (Figs 2, 3); in fact, in these views the ultrasonic

3 2D echo and CT in aortic aneurysm 547 beam is almost perpendicular to the anterior and the posterior aortic walls. Aortic arch: the arch is certainly the most difficult portion of the aorta to image; in our study it was visualized only from the suprastemal notch (Figs 1A, 4) the beam was directed inferiorly and posteriorly and slightly rotated in order to visualize the long axis of the ascending aorta, the arch and the upper portion of the descending thoracic aorta. Descending thoracic aorta: because of the length of the DTA different tomographic planes are needed to visualize it completely (Figs 1B-D, 5-7). The upper portion of the DTA was imaged from the suprastemal approach, as described. The central portion was visualized in a transverse or oblique section utilizing the precordial long axis and a series of continuous short axis views from the ascending aorta level to the papillary muscle levell7!. The DTA was imaged longitudinally, from the precordium, placing the transducer several centimetres from the left sternal border and direct- ing the plane of sweep in a superior-inferior direction, or from a modified apical approach, recently utilized in our laboratory (Figs 1C, 6, 7): the transducer was placed in the region of the apex and the sweep plane directed perpendicularly to the sternum in order to visualize the short axis of the DTA; subsequently the sweep plane was rotated 90 clockwise and directed medially to visualize the central and the lower part of the DTA longitudinally. The subcostal approach was also utilized to image the lower part of the DTAI'l: with the patient in the supine position, the transducer was placed in the subcostal position and rotated in order to visualize the abdominal aorta longitudinally: with a slight superior tilt of the transducer the inferior portion of the descending thoracic aorta was visualized (Figs 1D, 6C). An aneurysm was defined as a localized dilatation of the thoracic aorta. The diagnosis was made on a qualitative basis; however, normal values for the diameter of each aortic portion were Figure 1 Longitudinal echocardiographic views, and schematic diagrams, of the thoracic aorta. Panel A: Suprastemal echocardiography visualizes the ascending aorta and a large portion of the arch. Panel B: The initial portion of the descending thoracic aorta can also be visualized by the suprastemal approach. In panel C the middle and lower portion of the descending thoracic aorta is visualized with the transducer located in the proximity of the cardiac apex. Panel D: Subcostal visualization of the lower portion of the thoracic aorta.

4 548 S. lliceto et al. Figure 2 Precordial long axis view in a patient with a large aneurysm of the ascending aorta (Ao). LV = left ventricle. considered for comparison. In our laboratory, normal values of the diameter of each portion of the thoracic aorta were assessed by examining 20 normal volunteer subjects. The mean values obtained were: 26-3 mm for the ascending aorta, 23-9 mm for the aortic arch; 20-1 mm for the descending thoracic aorta. Care was taken in gain adjustment in order to obtain the best image resolution. 2D echo images were permanently recorded on a 1/2 inch video tape cassette for further analysis. COMPUTED TOMOGRAPHY Computed tomography of the chest was performed with a second generation type scanner (ACTA 2000 FS-Pfizer). The entire patient's chest was scanned with the patient in the supine position, the slices were contiguous and 13 mm thick, the scanning time of each slice was 18 s. Chest scanning was performed before and after intravenous injection of a bolus of 30 cc of contrast material (Urographin 75). SURGERY Patients 1 and 2 underwent surgery: in both cases the presence of a dissection involving the ascending aorta and the arch was confirmed. Results Two-dimensional echocardiography: the ascend-

5 2D echo and CT in aortic aneurysm 549 Figure 3 Precordial short axis, in diastole (upper panel) and systole (lower panel), of the aortic root in a patient with an aneurysm of the entire thoracic aorta. Semilunar valves (1-3) are well visualized both in diastole and systole. RA = right atrium; LA = left atrium. ing aorta was visualized in all patients; the aortic arch was not imaged in 4 cases, the DTA in 1 case. The entire thoracic aorta was completely visualized in 10 of the 14 patients. In case no. 1 an intimal flap dividing the true lumen from the false lumen of the aorta was detected by all approaches exploring the thoracic aorta (Fig. 8). In case no. 2 a flap was visualized in the ascending aorta and, only by the apical approach, in the DTA (Fig. 9). In case no. 3 the intimal flap was visualized in the DTA only by the suprasternal approach even though the descending aorta itself could be visualized by other tomographic planes.

6 /\ / Ao. ARCH \ \ Figure 4 Two-dimensional echocardiographic visualization, from the suprastemal approach, of the aortic arch in a patient with an aneurysm, without dissection, of the entire thoracic aorta. Figure 5 Two-dimensional echocardiographic visualization of the descending thoracic aorta (DTA) from the suprastemal approach in a patient with aneurysm, without dissection, of DTA.

7 2D echo and CT in aortic aneurysm 551 Figure 6 Patient no. I. Aortography (upper left panel) and two-dimensional echocardiographic examination (right panels A-C) of a patient with type 1 dissection. An intimal flap is clearly visualized in the thoracic aorta both by angiography (black arrows) and by 2D echo (white arrows). The entire thoracic aorta is visualized by 2D echo from the suprastemal (panel A), the apical (panel B) and subcostal approach (panel Q. DTA=descending thoracic aorta. Two-dimensional echocardiography failed to demonstrate the calcification of the aortic wall in case no. 7 and the thrombotic stratification in the DTA lumen in case no. 11. In two cases without dissection extra linear echoes were detected in the aortic lumen; in one of these cases, dissection was erroneously suspected because of the reproducibility of this finding; this linear echo was adjacent to the aortic wall and had a movement, during the cardiac cycle, that was parallel to the aortic walls. Computed tomography: an examination of satisfactory quality was obtained in all patients. In each case correct identification of the site and extension

8 552 S. Iliceto et al. Figure 7 Patient no. 2. Computed tomography (upper panel) and two-dimensional echocardiographic examination (lower panels) in a patient with a type I dissection. The intimal flap (arrows) is visualized by two-dimensional echocardiography both in the ascending (lower right panel) and in the descending thoracic aorta (DTA) (lower left panel), conversely computed tomography failed to demonstrate the dissection. AA = ascending aorta; TL=true lumen; La=left atrium. of the aneurysm was achieved. In none of the cases with dissection were intimal flaps detected. Both the calcification of the ascending aorta (Fig. 8) and the thrombotic stratification in the DTA lumen (Fig. 9) were clearly demonstrated. Discussion When an aortic aneurysm has been diagnosed, the changes of the aortic internal diameter must be followed over time because of the crucial relationship between the internal diameter of aortic aneurysms and their natural history: aneurysms that have a diameter larger than 6 cm are definitely more prone to rupture than smaller onesl'l. Moreover, a dissection may extend weeks to months after its onsett' J; therefore a non-invasive technique capable of recognizing these conditions in the early stages and of examining the entire thoracic aorta repeatedly would be welcome. Angiography is certainly the 'gold standard' for the diagnosis of this condition!') but because of its invasive nature it cannot be performed repeatedly in the same patient; furthermore the angiographic diagnosis of a dissection, based on the recognition of a double lumen and of an oscillating flap, is not always possible especially because the critical clinical status of these patients does not allow multiple angiographic projections to be performed since these require injections of large amount of contrast material. Two-dimensional echocardiography is noninvasive and can be performed repeatedly, if necessary at the patient's bedside; its ability in scan-

9 2D echo and CT in aortic aneurysm 553 Figure 8 Patient no. 7. Computed tomography shows diffuse calcification (arrows) of the walls of the ascending aorta (AA) and of the aortic arch. ning the entire thoracic aorta and in diagnosing aneurysm involving both the ascending^) and the descending aortal 4! has been assessed. De Maria el a/.p! correctly identified the site and extension of the dilatation in 12 patients with an aneurysm of the ascending aorta; furthermore, they showed an excellent correlation (r=0-91) between cineangiographic and echocardiographic values for aortic size. Victor el alw visualized an intimal flap in 12 of 15 patients with aortic dissection; only in one of the 27 patients without dissection did they have a false positive diagnosis. We reported the echocardiographic findings of 15 consecutive patients with aneurysms of the descending thoracic aortal' 4!: an intimal flap was identified in each case with dissection, whereas among the patients with aneurysms without dissection only one false positive was observed. Computed tomography is also a non-invasive technique but involves radiation and is much more expensive than 2D echo. Several studies have demonstrated its value in many kinds of cardiac diseases. Among our patients, the entire thoracic aorta was completely visualized in 71% of cases by 2D Echo and in all cases by CT. An intimal flap was correctly identified by 2D echo in the three cases with a dissection whereas in no case did CT allow such a recognition. We cannot exclude the possible presence of false negatives as

10 554 S. Iliceto et al. Figure 9 Patient no. 11. Computed tomography of the descending thoracic aorta (DTA) before the injection of contrast material. The thrombus is well visualized after contrast injection. The arrows indicates the lumen of DTA. we have considered angiography the 'gold standard' for this diagnosis. A dissection, in fact, can sometimes be missed by angiographic studies because of inadequate opacification of the false lumen or inappropriate angiographic projection. Both the calcification of the aortic walls and the thrombotic stratification of the DTA were missed by 2D echo, possibly because of inappropriate gain setting, but were correctly identified by CT. In two patients without dissection extra linear echoes parallel to the aortic walls were detected, these echoes were probably due to a stronger reflection of an atherosclerotic aortic wall. Our data suggest that 2D echo is more reliable than CT in diagnosing dissections of the aorta. However, it should be noted that it was not possible to obtain an adequate two-dimensional echocardiographic study of the entire thoracic aorta in all patients. It is well known, in fact, that not all tomographic planes are obtainable in all subjects because

11 2D echo and CT in aortic aneurysm 555 of many limiting factors! 8!; furthermore, as we have pointed out in a recent study, not all tomographic planes can visualize an intimal flap: in fact, because of its spiral nature, the flap can lie in a plane parallel or tangential to the echo beam so that an artifactual drop out of echoes may occur! 4!. It should also be noted that two-dimensional echocardiography was compared, in this study, with data obtained with a second generation CT scanner with an 18-s scanning time. This long scanning time can explain the lack of visualization of the intimal flap that moves rapidly during the whole cardiac cycle, thus causing an image overlapping that prevents its recognition. The newest CT scanners appear to be considerably better in the evaluation of dissections because of their faster scanning time (about 3 s); unfortunately, they are more expensive and in less widespread use than those of the second generation. In conclusion, according to our experience, 2D echo appears to be superior to conventional CT in * the recognition of an intimal flap due to a dissection because of its faster image displaying (30 images per s): conversely, CT allows the accurate recognition of the configuration and extent of the aneurysm and of the tissue modifications involving either the lumen or the walls of the aorta. t References [1] Arciniegas JC, Soto B, Little WC, Papapietro SE. Cineangiography in the diagnosis of aortic dissection. Am J Card 1981; 47: [2] De Maria A, Bommer W, Neumann A, Weinert L, Borgen H, Mason DT. Identification of aneurysm of ascending aorta by cross-sectional echocardiography. Circulation 1979; 59: [3] Victor MF, Mintz GS, Kotler MN, Wilson AR, Segal BL. Two-dimensional echocardiographic diagnosis of aortic dissection. Am J Cardiol 1981; 48: [4] Iliceto S, Antonelli G, Biasco G, Rizzon P. Twodimensional echocardiographic evaluation of aneurysms of the descending thoracic aorta. Circulation 1982; 66: [5] Godwin JD, Herfkens RL, Skioldebrand CG, Federle MP, Lipton MJ. Evaluation of dissections and aneurysm of the thoracic aorta by conventional and dynamic CT scanning. Radiology 1980; 136: [6] Gross SC, Barr I, Eyler WR, Khaja F, Goldstein S. Computed tomography in dissection of the thoracic aorta. Radiology 1980; 136: [7] Mintz GS, Kotler MN, Segal BL, Parry WR. Twodimensional echocardiographic recognition of the descending thoracic aorta. Am J Cardiol 1979; 44: [8] Bansal RC, Tajik AJ, Offord KP. Feasibility of detailed two-dimensional echocardiographic examination in adults. Mayo Clin Proc 1980; 55: 291. [9] Braunwald E. Heart disease. Philadelphia: W.B. Saunders, 1980: [10] Lindsay J Jr, Hurst WJ. Clinical features and prognosis in dissecting aneurysm of aorta. Circulation 1967; 35: